© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 15996
    Prev
    Next
    INDEX
    Explanations

    of, to, for

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on tokens carrying quantitative or measurement information (numbers and the words immediately around them), i.e. it’s looking for numerical data or metrics.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     after
    -2.28
     in
    -2.14
     at
    -2.00
     on
    -1.94
     for
    -1.93
     should
    -1.92
     by
    -1.85
     with
    -1.82
     so
    -1.80
     of
    -1.79
    POSITIVE LOGITS
     and
    2.14
    และ
    1.59
     både
    1.55
     และ
    1.52
     vijf
    1.47
     laranja
    1.45
     abd
    1.43
     LITER
    1.40
     certamente
    1.40
     véhicules
    1.40
    Activations Density 0.177%

    No Known Activations